The new exported DSU status removes the need to make blocking binder
calls out of system server during device boot.
Bug: 149790245
Bug: 149716497
Test: adb shell am start-activity \
-n com.android.dynsystem/com.android.dynsystem.VerificationActivity \
-a android.os.image.action.START_INSTALL \
-d file:///storage/emulated/0/Download/system.raw.gz \
--el KEY_SYSTEM_SIZE $(du -b system.raw|cut -f1) \
--el KEY_USERDATA_SIZE 8589934592
Change-Id: I27fae316214498407a73474ca8b93aec3518e4b5
Merged-In: I27fae316214498407a73474ca8b93aec3518e4b5
(cherry picked from commit 2e1c6b26a1)
A previous change moved property_service into its own thread, since
there was otherwise a deadlock whenever a process called by init would
try to set a property. This new thread, however, would send a message
via a blocking socket to init for each property that it received,
since init may need to take action depending on which property it is.
Unfortunately, this means that the deadlock is still possible, the
only difference is the socket's buffer must be filled before init deadlocks.
There are possible partial solutions here: the socket's buffer may be
increased or property_service may only send messages for the
properties that init will take action on, however all of these
solutions still lead to eventual deadlock. The only complete solution
is to handle these messages asynchronously.
This change, therefore, adds the following:
1) A lock for instructing init to reboot
2) A lock for waiting on properties
3) A lock for queueing new properties
4) A lock for any actions with ServiceList or any Services, enforced
through thread annotations, particularly since this code was not
designed with the intention of being multi-threaded.
Bug: 146877356
Bug: 148236233
Test: boot
Test: kill hwservicemanager without deadlock
Merged-In: I84108e54217866205a48c45e8b59355012c32ea8
Change-Id: I84108e54217866205a48c45e8b59355012c32ea8
(cherry picked from commit 7205c62933)
Users of libinit_test_utils must include all libraries that it uses.
If it uses libinit, then there is a large number of libraries that
must be included. To avoid this, make libinit_test_utils only use
init_common_sources and the small number of required libraries that go
along with those sources. Additionally, expose these sources as a
default for users of libinit_test_utils.
Bug: 148236233
Test: build
Merged-In: I224fa7e0590d073e4cd40412b5dcb6f72a64b6bf
Change-Id: I224fa7e0590d073e4cd40412b5dcb6f72a64b6bf
(cherry picked from commit a2f9136b2c)
If init is wedged, then the write will never succeed and reboot won't
happen.
Also, in case of normal reboot, move call to PersistRebootReason to the
top of DoReboot() function, to make sure we persist it even if /data is
not mounted.
Test: builds
Test: adb shell svc power reboot userspace
Test: atest CtsUserspaceRebootHostSideTestCases
Bug: 148767783
Change-Id: I4ae40e1f6fdc41cc0bcae57020fa3d3385dda1b4
Merged-In: I4ae40e1f6fdc41cc0bcae57020fa3d3385dda1b4
Prevent appending ",adb" to persist.sys.usb.config if "none" is
explicitly defined in default prop.
Bug: 150130503
Test: persist.sys.usb.config initalized correctly
Change-Id: I3b5de6fd102e252019e843f39f0875f5aaea7486
Helps with support of recovery and rollback boot reason history, by
also using /metadata/bootstat/persist.sys.boot.reason to file the
reboot reason.
Test: manual
Bug: 129007837
Change-Id: Id1d21c404067414847bef14a0c43f70cafe1a3e2
Those keys will be used for key revocation check by DSU installation
service. Note that failing to copy those keys to /metadata is NOT fatal,
because it is auxiliary to perform public key matching prior to booting
into DSU images on next boot. The actual key matching will still be done
on next DSU boot.
Bug: 146910547
Test: boot device, checks the avb keys are copied to /metadata/gsi/dsu/avb/.
Change-Id: I25a4eba82e84288bac7a859205c920628a063651
These can be AIDL or HIDL interfaces. Avoiding assuming the format here.
Bug: 141008902
Test: vts_ibase_test
Change-Id: If60367e3a46c74a1d28548379655f5e211f58b73
Instead they will be logged from system_server. This CL just prepares
grounds for logging CL to land.
Test: adb reboot userspace
Bug: 148767783
Change-Id: Ie9482ef735344ecfb0de8a37785d314a3c0417ff
fmtlib provides compile time checking of format strings that we're not
currently using. This change makes Errorf() and ErrnoErrorf() into
macros such that we can take advantage of this capability.
Test: build successfully normally
Test: fail the build if using an invalid format string
Change-Id: Icb8ba8cb973bbd1fa4755a62e7598bdbb0113757
Avb keys used to verify a partition are stored in the first-stage
ramdisk. However, after /system is mounted, init will chroot into it.
This makes those keys inaccessible for later mounts, e.g., /vendor or
/product.
This change retains avb keys by reading all of them before chroot
into /system.
Note that it is intentional to perform public matching for both
preload_avb_key_blobs and fstab_entry.avb_keys in libfs_avb.
As some keys might only be availble before init chroots into /system,
e.g., /avb/key1 in the first-stage ramdisk, while other keys might
only be available after the chroot, e.g., /system/etc/avb/key2.
Bug: 147585411
Test: specify avb_keys for a partition and checks the keys are preloaded
Test: atest libfs_avb_test
Test: atest libfs_avb_internal_test
Change-Id: I6bd490c4215480db2937cdfc3fea0d616e224a91
These operators were included because they're present in the draft
standard proposal of std::expected, but they were deemed to lead to
bugs, particularly when T is implicitly convertible to bool.
Change-Id: Ib149decf1f230198f358dc1ae0eaed71961363f6
Test: m
We currently do not handle process actions (restarting services or
exiting timedout services) when we are waiting for an exec service,
but this seems to be the wrong behavior. Particularly, an exec
service may depend on a previously started service and if that service
crashes, we will deadlock unless init restarts it.
Bug: 146920034
Test: build, boot
Change-Id: Id2fc936b8a7b989862ba4c32c398a544941e0e76
Currently, this is only enabled for Virtual A/B devices. When /data uses
F2FS, then proper pinning support must be enabled.
Because /data cannot be mounted in recovery, we can't delete the scratch
image backed by /data while in fastbootd. Instead, we mark it as
disabled in /metadata. The remount command now has an init script that
checks for and removes a disabled scratch partition.
Bug: 134949511
Test: adb remount on V A/B device with patched F2FS
Change-Id: Ifc8720378259654472d3822e97059b6c366f601d
Also reset some more properties to make bootanimation work properly.
Test: adb reboot userspace
Bug: 148172262
Change-Id: I0154d4fe9377c019150f5b1a709c406925db584d
mkdir always returns -1 for any types of errors.
errno should be checked for actual error type.
Test: m
Change-Id: I1b56d48ba48992a2f9629dc09d795c277b5b774d
This gives search (x) permission on the parent apex data
directory /data/misc/apexdata so that directories below it
may be opened. It also gives that permission on the apex
data directories themselves.
Bug: 147848983
Test: Build & flash, check perms are correct
Change-Id: I27c4ea01602002c89d0771a144265e3879d9041a
Historically, the syscall was controlled by a system-wide
perf_event_paranoid sysctl, which is not flexible enough to allow only
specific processes to use the syscall. However, SELinux support for the
syscall has been upstreamed recently[1] (and is being backported to
Android R release common kernels).
[1] da97e18458
As the presence of these hooks is not guaranteed on all Android R
platforms (since we support upgrades while keeping an older kernel), we
need to test for the feature dynamically. The LSM hooks themselves have
no way of being detected directly, so we instead test for their effects,
so we perform several syscalls, and look for a specific success/failure
combination, corresponding to the platform's SELinux policy.
If hooks are detected, perf_event_paranoid is set to -1 (unrestricted),
as the SELinux policy is then sufficient to control access.
This is done within init for several reasons:
* CAP_SYS_ADMIN side-steps perf_event_paranoid, so the tests can be done
if non-root users aren't allowed to use the syscall (the default).
* init is already the setter of the paranoid value (see init.rc), which
is also a privileged operation.
* the test itself is simple (couple of syscalls), so having a dedicated
test binary/domain felt excessive.
I decided to go through a new sysprop (set by a builtin test in
second-stage init), and keeping the actuation in init.rc. We can change
it to an immediate write to the paranoid value if a use-case comes up
that requires the decision to be made earlier in the init sequence.
Bug: 137092007
Change-Id: Ib13a31fee896f17a28910d993df57168a83a4b3d
In case one of the cryptfs calls to vdc fails, first try a normal
reboot.
Test: manual
Bug: 135984674
Bug: 143970043
Change-Id: I5706d9cd6c3a08fa06329ffb7d141de632620e3d
Scoped storage has some unique requirements that are hard to implement
with the two mount namespaces, because the daemon that does the mounting
(vold) lives in a different namespace than the processes using those
mounts.
In particular, /mnt/installer is a special bind mount that should
receive mount events under /mnt/user, but at the same time only
only propagate mount events under /mnt/installer to /mnt/installer in
the other namespace. More details in the code.
Bug: 134706060
Test: /mnt/installer shows up and is setup correctly.
Change-Id: I6dab5ace5a345d9d684a9f1ae94c833fc294d49e
* Open the ImageManager according to the saved DSU slot.
* Load the saved install_dir when it presents. This is required
when installing to an external storage and the current code
seems to break it.
* Remove an unnecessary C++ using.
Test: gsi_tool install
Bug: 144247097
Change-Id: I5b7be25787f235518a25abacefa612c33861f6ef
While handling the fatal signals and abort, exits from the signal
handler to panic the system if "androidboot.init_fatal_panic" in cmdline
is configured as "true".
Bug: 146818493
Change-Id: I59fffb7598ce981383ae24961a97fd2fd8e8d64e
Commit Ida1a6da988c10d364b3ccdaa6c5d63e5264d1b27 allows specifying
AVB-related partitions to be probed in first-state mount, e.g.,
"avb=vbmeta_system" in the fs_mgr_flags will make init wait uevents
for vbmeta_system partition, which is required for AVB verification.
However, when specifying "avb=dtbo" to chain a /dtbo partition, first-stage
init will emit the following error as it assumes all partitions are
verified by Hashtree descriptors.
init: [libfs_avb]Hashtree descriptor not found: dtbo
This change fixes this by skipping raw partitions for first-stage mounting,
where raw partitions should already be verified by bootloader with Hash
descriptors. Although first-stage init does not perform Hash verification,
it still reads all Hash/Hashtree descriptors to calculate then compare
the "VBMeta Digest" passed from bootloader.
https://android.googlesource.com/platform/external/avb/#The-VBMeta-Digest
The following is an example of a fstab entry to chain /dtbo partition:
/dev/block/bootdevice/by-name/dtbo /dtbo emmc defaults slotselect,avb=dtbo,first_stage_mount
Bug: 146935233
Test: enable /dtbo as a chain partition, then boot the device
Change-Id: I590f1d9b78ef89b5aef21cb7bf19fdb90b27f808
Marks this CTS test module as supporting secondary_user, meaning that
the tests are eligible to be run from a regular Android user other
than the system user (i.e. other than user 0).
Note that 'user' here refers to the concept of users in Android
Multiuser; it is NOT the same as a uid.
Bug: 141773058
Test: module passed with secondary_user params
(--enable-optional-parameterization --module-parameter secondary_user)
Change-Id: Ic8d29cabe2542d75c8f873b677550db9f11b3050
The apex data directories must be accessed by apexd in order for it to
perform snapshot and restore as part of the rollback process. As apexd
runs as root, this CL changes the apex data directories under /data/misc
to be owned by root.
Bug: 141148175
Test: Build and flash; check permissions are set correctly.
Change-Id: Ib534e705802c06900884a15f39fee257d4987f4c
When BCB (bootloader message structure inside of misc partition) is
malformed (contains some non-printable characters in its fields),
"reboot recovery" command won't be able to write required string to
"command" field. It can happen for example when partition table was
created anew and 'misc' partition area contains some garbage. Also this
behavior can be emulated with this command:
$ fastboot erase misc
which leads to 'misc' partition to be filled with 0xFF characters. Hence
this code:
if (boot.command[0] == '\0') {
won't let us to set new string to "command" field. Let's check if
"command" field is malformed and fix it, before actually checking for
previously set content.
"fastboot erase" shouldn't be used for testing purposes though, as it
doesn't work sometimes due to alignment, on bootloader side:
Erasing blocks 6144 to 6144 due to alignment
........ erased 0 bytes from 'misc'
Instead one might use "dd" command to fill 'misc' with 0xFF's:
$ dd if=/dev/zero ibs=2k count=1 | tr "\000" "\377" >misc.img
$ fastboot flash misc misc.img
Test: Fill 'misc' partition with 0xFF's, then do "adb reboot recovery"
Change-Id: Ica8ca31012b9b2249645e7305830c07a20dd013c
Signed-off-by: Sam Protsenko <semen.protsenko@linaro.org>
Current linkerconfig targets for specific output file. However,
linkerconfig will generate more than 1 file based on APEX modules, so it
should take argument for target directory rather than target file. This
change updates linkerconfig's argument to point output directory.
Bug: 146993126
Test: m -j passed & Cuttlefish succeeded to boot
Change-Id: I3a720a047077688582436aabd307adafeafc5398
Since I was there, added two more properties to reset, and switched
ordering of sys.init.updatable_crashing and
sys.init.updatable_crashing_process_name setprops to make sure that
process name is already set when apexd/PackageWatchdog get's notified
about sys.init.updatable_crashing.
Also fixed a typo in what HandleUserspaceReboot function.
Test: adb reboot userspace
Bug: 135984674
Change-Id: I954ec49aae0734cda1bd833ad68f386ecd808f73
There are two namespaces from init - bootstrap and default - and those
will have different set of APEX modules. To support difference between
two namespaces, linker config should be generated per namespace and each
namespace should use its own linker configuration. As a first step of
the work, this change will create different mount point for each
namespace, and re-generate linker config after APEX mount from each
namespaces.
Bug: 144664390
Test: m -j passed & tested from cuttlefish
Change-Id: Iac2e222376ec4b0ced6c29eed18b21d39ff0b1ba
The first_stage_mount code for DSUs predates both the DSU name and the
ImageManager abstraction. Move this code to ImageManager, so it can be
shared with overlayfs/scratch handling. And while we're here, rename GSI
to DSU for clarity.
Bug: 134949511
Test: dsu works
Change-Id: I9ef374bccc6cdbe2ada88baef4e7c0bc81b1e85e
When bind-mounting flattened APEX, use the apex name found in
apex_manifest.pb as the mount point, instead of the directory name which
might be different from apex name in case when the apex is overridden.
This allowed us to remove the special casing for the ART apex since we
/system/apex/com.android.art.release will be mounted to
/apex/com.android.art instead of /apex/com.android.art.release.
Bug: N/A
Test: m
Test: OVERRIDE_TARGET_FLATTEN_APEX=true m, device is bootable
Change-Id: Ibdde7002b9078db390e6672b0eb82c474925451d
The previous code would lazily check for 'exact' and accept any other
value as a prefix match. This should be a tighter check allowing only
'exact', 'prefix', or an empty string for this option.
Test: build fails if an invalid string is used for the match operation
Test: build succeeds normally
Test: `getprop -Z` shows exact vs prefix is differentiated correctly
Change-Id: I21dcb193810d65f468f8960967eabfd261f71e21
As part of Watchdog-triggered rollback logging, it would
be useful to know the service which has caused
updatable_crashing to be set. Store this in a new property
sys.init.updatable_crashing_process_name.
Test: manual
Bug: 138782888
Change-Id: Iffa38b2b746abbb5d823a8d69b7b8d2456836f4e
We do not need to set up all directories from system.img for the first
stage ramdisk, only those which first stage init mounts on.
Test: n/a
Change-Id: I877c623bac0b6dd2927ebdefa20da929ff98b341
This creates the directory /data/misc/apexdata, at the same time as other
directories under /data/misc. Then, when apexd has finished activating
APEXes, a directory is created under /data/misc/apexdata for every
APEX, with the same name as the APEX module name.
See go/apex-data-directories.
APEXes are discovered by scanning the /apex directory. It may be better
to delegate this process to a library, but it is proposed to defer that
change to a future CL.
Bug: 141148175
Test: Built and flashed, checked directories were created.
Change-Id: I639d6f490ae0b97f116ce38ff3ac348bd73aa20e
Currently linker config locates under /dev, but this makes some problem
in case of using two system partitions with chroot. To match system
image and configuration, linker config better stays under /linkerconfig
Bug: 144966380
Test: m -j passed && tested from cuttelfish
Change-Id: Iaae5af65721eee8106311c1efb4760a9db13564a
Such services will be re-parsed and added back to the service list
during post-fs-data stage.
Test: adb reboot userspace
Test: atest CtsInitTestCases
Bug: 145669993
Bug: 135984674
Change-Id: Ibb393dfe0f101c4ebe37bc763733fd5d981d3691
Init is no longer a special case and talks to property service just
like every other client, therefore move it away from property_set()
and to android::base::SetProperty().
In doing so, this change moves the initial property set up from the
kernel command line and property files directly into PropertyInit().
This makes the responsibilities between init and property services
more clear.
Test: boot, unit test cases
Change-Id: I36b8c83e845d887f1b203355c2391ec123c3d05f
During userspace reboot FscryptInstallKeyring will be called again, this
CL will make it second call a no-op, which IMHO is better than having a
special logic in init to conditionally call FscryptInstallKeyring
depending on whenever it's normal boot, or userspace reboot.
Test: adb reboot userspace
Test: checked in kernel logs that new keyring is not created
Bug: 135984674
Change-Id: I4ad5aee6887b7318fb1cd02bf1c7be8da6ece599
There are a few directories for builtin APEXes: /system/apex,
/system_ext/apex, /product/apex, /vendor/apex.
For devices with ro.apex.updatable=false, init should bind-mount
"flattened" apexes from all of above.
Bug: 144732372
Test: OVERRIDE_TARGET_FLATTEN_APEX=true m && device boot
Change-Id: If151145e6e3d966b202300ff3a0ddb5ec9fdaa87
A recently added subcontext test was failing beause it was running as
non-root, but GTEST_SKIP() didn't work as I expected it to.
In retrospect, all of these tests except for the property one, can
easily run as root, so this changes allows all of these tests to run
as root, while fixing the original issue.
Bug: 144707143
Test: root and nonroot subcontext unit tests
Change-Id: Ia835597701698f6be2101f92d6f4c9450bd3c7dd
~2007 a change was added that would allow oneshot services to
daemonize by not killing their process group. This was a hack at the
time, and should certainly not be needed now. I've resisted removing
the behavior however, as it hadn't caused any issues.
Recently, it was detected that the cgroups that these processes belong
to, would exist forever and therefore leak memory. Instead of simply
removing the cgroups when empty, this provides a good opportunity to
do the right thing and fix this behavior once and for all.
The new (correct) behavior only happens for devices with vendor images
built for Android R or later. Init will log a warning to dmesg when
it detects this difference in behavior has occurred.
Bug: 144545923
Test: boot CF/Coral and see no difference in behavior.
Test: boot CF with a service that daemonizes and see the warning.
Change-Id: I333a2e25a541ec0114ac50ab8ae7f1ea3f055447
sys.init.userspace_reboot.in_progress will be used to notify all
the processes (including vendor ones) that userspace reboot is
happening, hence it should be treated as stable public api.
All other sys.init.userspace_reboot.* props will be internal to /system
partition and don't require any stability guarantees.
Test: builds
Test: adb reboot userspace
Bug: 135984674
Change-Id: Ifb64a6bfae2de76bac67edea68df44e33c9cfe2d
Watchdog is just a forked process that is going to fall back to the
full reboot in case device wasn't able to boot in given amount of time.
Currently this amount is hard-coded to 1 minute, but in the future it
will be controlled by a read-only property.
Also added sync calls before and after tearing down services.
Test: adb reboot userspace
Bug: 135984674
Change-Id: Ie6053c9446a6761deae6dc104036bb35b09ef0e2
Previously CheckPropertyTriggers() tried to do the entire property
triggers check with one loop. However, that would require calling
GetProperty() on all properties for all triggers just in case the
property that is being set is used by a given trigger.
This change first checks that the property being set exists in each
trigger and that its value is set such that the trigger would be
triggered, only then does it check that other property triggers are
set to the right value.
Bug: 143922756
Test: boot
Test: substantially fewer GetProperty() calls from
CheckPropertyTriggers()
Change-Id: I0228cf47328b31963eaf3fc689fb60f711532df4
There will be useful in debugging/logging events to statsd.
Also as part of this CL, sys.init.userspace_reboot.in_progress property
is now used as a mean of synchronization. It is set directly in
DoUserspaceReboot, to make sure that all the setprop actions triggered
by userspace-reboot-requested were processed.
Test: adb reboot userspace
Test: adb shell getprop sys.init.userspace_reboot.last_started
Test: adb shell getprop sys.init.userspace_reboot.last_finished
Bug: 135984674
Change-Id: I9debcd4f058e790855200d5295344dafb30e496a
Previously, we assumed that TriggerShutdown() should never be called
from vendor_init and used property service as a back up in case it
ever did. We have since then found out that vendor_init may indeed
call TriggerShutdown() and we want to make it just as strict as it is
in init, wherein it will immediately start the shutdown sequence
without executing any further commands.
Test: init unit tests, trigger shuttdown from init and vendor_init
Change-Id: I1f44dae801a28269eb8127879a8b7d6adff6f353
Use TEMP_FAILURE_RETRY() for recvmsg() to prevent failing
unnecessarily from EINTR. Use PLOG(ERROR) in ueventd to display the
actual error if reading from the uevent socket does fail.
Test: build
Change-Id: I64a22698ce6ffc7573f75b57803551c7b6940e30
Howard Chen